{"id":"W2909244445","doi":"10.1145/3278721.3278728","title":"Towards Provably Moral AI Agents in Bottom-up Learning Frameworks","year":2018,"lang":"en","type":"article","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Top-down and bottom-up design; Artificial intelligence; Cognitive science; Programming language; Psychology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01567441,0.001288305,0.001246405,0.001414686,0.002099861,0.005110681,0.003279948,0.004241579,0.004233931],"category_scores_gemma":[0.04904649,0.0008141545,0.001443682,0.0007781335,0.01119439,0.006615051,0.009155268,0.006503243,0.0007996171],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003184231,"about_ca_system_score_gemma":0.002630816,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00142078,"about_ca_topic_score_gemma":0.00146355,"domain_scores_codex":[0.9891509,0.00653645,0.0003359366,0.001203351,0.002071485,0.000701917],"domain_scores_gemma":[0.9680233,0.02183299,0.002338304,0.004293746,0.002363546,0.001148129],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00002906314,0.00003325069,0.0003428085,0.00006261203,0.00003760147,0.00007521997,0.0003834596,0.06528025,0.0004574484,0.9264922,0.0005872163,0.006218803],"study_design_scores_gemma":[0.00001358361,0.00002013406,0.00004127114,0.00002146483,0.000009922468,0.00001412749,0.00004672856,0.1670712,0.0004086887,0.8311692,0.001171899,0.00001169059],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01834635,0.0001959054,0.9643753,0.003390435,0.00006774145,0.0000847897,0.00005137666,0.0003338608,0.01315431],"genre_scores_gemma":[0.7139131,0.0002397793,0.2779277,0.001211734,0.0001728606,0.0003550981,0.0001049363,0.0002424661,0.005832323],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01567441,"threshold_uncertainty_score":0.08289522,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02601395670154006,"score_gpt":0.3166067087152982,"score_spread":0.2905927520137582,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}